Decision Making in Uncertain Times: An interview with Donald Hantula, Ph.D.

Decisions made in organizations can seem baffling, even infuriating, especially during times of downturn and uncertainty. John Thain, CEO of Merrill Lynch, for example, recently gave away billions in bonuses and redecorated his office to the tune of $1.22 million while his failing company was being absorbed in a takeover funded by U. S. taxpayers. GM poured money into Saturn for 24 years despite the fact that it never turned a profit. Attempting to understand why people make such seemingly irrational decisions, we interviewed leading behavioral researcher, Don Hantula, Ph.D., director of the Decision Making Research Laboratory at Philadelphia’s Temple University.

As a significant contributor to the study of decision making, Hantula points out two critical areas of decision making in his research: escalation of commitment (“throwing good money after bad” referring to resources already invested to justify continued investment, also known as “sunk costs”) and decision dilemma theory (how decisions are made when feedback on performance and desired results is unclear).

 

We asked Dr. Hantula about his research into decision dilemma theory and escalation of commitment and what managers and organization leaders can learn from such research.

How would you summarize escalation of commitment?

Escalation of commitment occurs when people recommit resources to a chosen course of action even when that course of action appears to be in conditions of failure.

What is decision dilemma theory?

Decision dilemma theory says that when people are in a situation where the outcome is not clear what people are doing is not making mistakes—people are adapting. Hindsight is always 20/20. You can look back and say, “Yes, we did or did not do that right” but in the midst of a very confusing situation, managers try to adapt the best they can. Adaptation doesn’t always mean that they succeed, but it does mean that they are trying to succeed. So don’t blame people for bad outcomes when the outcome isn’t well known or predictable. If you want people to be accountable, hold them accountable for the process. 

So, it is important to look at the person who is making the decision and consider all the variables acting on that person at the time?

Yes. Right now you have a lot of executives who are wondering, “What do I do?” because the economy in general looks pretty much like a down slope. However, not every company in every sector is in a decline, so executives are asking, “Do I abandon what I am doing?” A. J. Dixit in “Thinking Strategically” has a theory that whenever a decision maker is in a situation in which the outcomes or probabilities cannot be fixed, the best thing for that person to do is stay in the situation because, if nothing else, he or she is gathering information by doing so. Unless you have very clear information about failure or you have a situation in which your salvage costs are very high (if you cash out now you’ll make a lot of money) you should use the opportunity to gather more information. This suggests that when people are encountering failure [and it looks like they are making bad decisions] they may actually be gathering information. In previous research I showed that when confronted with an uncertain situation people will buy information even when that information is completely useless. People adapt to failure by gathering more information.

I can imagine that many successful people have been repeatedly reinforced for gathering information.

That’s part of it. The other part is explained in my article, “When Success Breeds Failure” which describes what happened when college students made marketing decisions in a pharmaceutical sales situation. They were asked to decide how to allocate money to marketing campaigns, and then received feedback about the successful impact of their marketing campaigns on sales. We ran one session with a group of students during which we designed it so that most of their decisions were profitable. Then we ran the first group and a second new group through an identical scenario, but this time we designed it so that their decisions led mostly to failure. We saw that the group with prior profitable experience persisted in throwing money into the failing situation much longer than did the new group. If we look at this from an evolutionary perspective it makes sense. Most people if they are to succeed in life and reach some decent level of employment have trained under some pretty tricky and changing reinforcement situations. So you would expect that when faced with some sort of downturn or failure they are going to keep at the behaviors that brought reinforcement and reward in the past.

Where has the research gone lately? What are some current findings?

We recently looked at group dynamics in an escalation of commitment situation. You and I both know that many major decisions in organizations are not made by individuals. They are made by teams. We looked at business students some of whom had a history of working together in their classes, and a second group that did not. We found that group cohesion did not factor into the results, but that when decisions were made in groups, escalation of commitment persisted more than when individuals made decisions. A second measure was that of conflict. We measured what people would decide individually and we then compared that to their group decision. We found that when individuals differed more from the group decision there was more escalation.

How did the clarity of feedback impact decision making?

The other manipulation we threw in was a standard by which to evaluate feedback. That is a tool for sense-making. Without a standard, people are confused. With a standard or any other rule for making sense of data, people can determine success or failure. 

 

Deming’s [see sidebar] genius was the idea that everything is variable and rather than look at—in a manufacturing example—output that is going up and down rapidly, just get your mean and standard deviation. If your output is within one or two standard deviations, you’re still fine. However, if you get to three standard deviations, then you’ve got a problem. That concept allowed people who were running machines to take a look at the data and ask, “Okay, is my machine off kilter?”

So what does this tell us we should do when feedback is highly equivocal?

That’s when the whole idea of standards becomes very important. If you have standards, methods, or rules for evaluating feedback you can simplify the feedback and you do not get caught in its variability. By definition the more simple feedback is, the less equivocal it is. It tells you success or failure. It is simple and takes out unnecessary complications.

So when delivering feedback we should consider the other person’s point of view and put the feedback in language meaningful to that individual. We should examine their world and how their tasks are defined and put the feedback in language that decreases the complexity.

Correct.

What advice would you give to managers and executives based on the research?

Focus on the process not the outcome.

If you hold people accountable for an outcome, they will do all sorts of things to get it, some of which may often be quite bad. Enron was a perfect example of this. In Enron’s “rank and yank” performance appraisal process, employees were ranked on a 1-5 scale every six months and the lowest ranked were terminated. Because many jobs at Enron involved ‘making numbers’ or trading and profit outcomes, many people now blame the culture this created—one that reinforced individual gain and cutthroat competition rather than teamwork, and that very likely reinforced some fudging of semi-annual numbers. But, if you have a very uncertain environment, you want people to make decisions based on processes that are established and accepted by the organization. The more uncertain an environment is, the more outcomes may well be random or due to luck—and holding people responsible for chance and luck is an excellent way to create a very dysfunctional organization. It is one thing to hold people responsible for not doing the things they were supposed to do. It’s another to hold them responsible for an outcome when they did do what they were supposed to do. In the first situation you are reinforcing rule following. In the second you are punishing rule following. In fact, if you have organization-established and accepted processes for creative problem solving and development it is best to reinforce following those processes because the outcomes are not always successful.

Is there anything else you would like to recommend to managers and executives?

In real downturn situations like the one we are in now, this is the best time to take risks.

The situation is such that you might sink no matter what; especially if you keep doing what you are doing. Then it is probable you will go under. Now, what you can do is either hope for dumb luck to turn in your favor or take big risks and reinvent yourself. If you are going under, do you go under slowly or do you try to do something that makes a difference?

Unfortunately most people during a downturn do the exact opposite—hold everything close and don’t take risks and just sink slowly.

I’ve heard that in times like these when most of your competitors are doing just that— holding close and not taking risks—now is the time to develop new products and services, reach out for new business, and make investments because it sets up a contrast between you and your competition.

Also, if the environment is going down quickly that will take care of most of your competitors anyway, so this is the time to take risks. 

Although we can’t know for sure what was going through the mind of John Thain as he handed out billions of dollars in bonuses, this excerpt from a memo he sent out a short time after doing so may provide some clues:

I want to address several topics that have been inaccurately reported in the press.  The first issue is our year-end bonus payments.  Our 2008 discretionary bonus pool was 41 percent lower than 2007.  The size of the pool, its composition (cash and stock mix), and the timing of the payments for both the cash and stock were all determined together with Bank of America and approved by our Management Development and Compensation Committee and our Board.  The total bonus pool was also substantially less than the amount allowed under our merger agreement.

 

All of his years at the top in financial institutions during boom times taught Thain that bonuses led to success. As in the studies described above, when negative feedback showed up in the form of a downturn, Thain, persisted in doing the same thing—delivering bonuses, just like in any other year.

Making the decision even easier were the processes set up within the organization and within the current takeover by Bank of America. Thain played by previously established rules, did what had brought him success before, and met his employees’ expectations. With all of these factors acting on him, it is a bit easier to understand (but not approve of) his behavior.

 

The auto executives also fell prey to some version of these decision-making phenomena. They all had long histories of success. When the car market started to decline, they continued to operate in the same manner. They continued to throw good money after bad. It took an unequivocal wake-up call involving severe consequences to get them to change their behavior.

 

The decision-making failures of John Thain, the U.S. auto execs, and many more like them highlight a well-known, but not completely understood platitude of success in most organizations: “It’s lonely at the top.” The higher that people climb in most organizations, the less candid feedback they receive. An individual becomes more and more responsible for his or her own feedback when surrounded by fewer and fewer people who are willing to say, “Uh, that sounds like a bad idea, boss.”

 

In times of downturn and uncertainty we need to constantly review our processes and the feedback they provide (or fail to provide). As people encounter failure and uncertainty, they will often adapt by doing the same things that proved successful in the past (throwing good money after bad) or by staying the course while attempting to gather helpful information. As always, but especially today, managers and executives must ensure that they and their associates are continually provided with the best, most relevant information and the right reinforcement for making the best choices at all times.

 

About Donald Hantula, Ph.D.

 

Dr. Hantula’s research expertise includes organizational behavior, evolutionary behavioral economics, occupational health and safety, managerial and consumer decision making, computer applications, and behavior analysis in organizational settings. Don is the former Executive Editor of the Journal of Social Psychology and his research has been featured in such prestigious publications as the Journal of Applied Psychology, Journal of Economic Psychology, the Journal of Organizational Behavior Management, Human Resource Management, Psychology and Marketing, and the Journal of Applied Behavior Analysis.

 

W. Edwards Deming

 

W. E. Deming was a pioneer in applying scientific and statistical methods to manufacturing processes. His methods revolutionized manufacturing and business principally in Japan and the United States starting in the 1950s and continuing until the 1980s. His teachings live on in modern manufacturing and business as the basis for such widely-used methods as Six Sigma, Lean Manufacturing, and Total Quality Management.

 

References

Ross, J., & Staw, B. M. (1986). Expo 86: An Escalation Prototype. Administrative Science Quarterly. 31(2), 274-97.

Drummond, H. (1996). Escalation in Decision-Making: The Tragedy of Taurus New York: Oxford University Press.

Dixit, A. J., & Nalebuff, B. J. (1991) Thinking strategically : the competitive edge in business, politics, and everyday life. New York: W.W. Norton & Co.

DeNicolis-Bragger, J. L., Hantula, D. A., Bragger, D., Kirnan, J., & Kutcher, E. (2003). When Success Breeds Failure: History, Hysteresis, and Delayed Exit Decisions. Journal of Applied Psychology, 88, 6-14.